Semantic Fault Localization and Suspiciousness Ranking
Static program analyzers are increasingly effective in checking correctness properties of programs and reporting any errors found, often in the form of error traces. However, developers still spend a significant amount of time on debugging. This involves processing long error traces in an effort to localize a bug to a relatively small part of the program and to identify its cause. In this paper, we present a technique for automated fault localization that, given a program and an error trace, efficiently narrows down the cause of the error to a few statements. These statements are then ranked in terms of their suspiciousness. Our technique relies only on the semantics of the given program and does not require any test cases or user guidance. In experiments on a set of C benchmarks, we show that our technique is effective in quickly isolating the cause of error while out-performing other state-of-the-art fault-localization techniques.
Mon 8 AprDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
14:00 - 16:00 | |||
14:00 30mTalk | LCV: A Verification Tool for Linear Controller Software TACAS Junkil Park University of Pennsylvania, Miroslav Pajic Duke University, Oleg Sokolsky University of Pennsylvania, USA, Insup Lee Link to publication | ||
14:30 30mTalk | Semantic Fault Localization and Suspiciousness Ranking TACAS Maria Christakis MPI-SWS, Matthias Heizmann University of Freiburg, Muhammad Numair Mansur Max Planck Institute for Software Systems (MPI-SWS), Christian Schilling IST Austria, Valentin Wüstholz ConsenSys Diligence Link to publication | ||
15:00 30mTalk | Computing Coupled Similarity TACAS Link to publication | ||
15:30 30mTalk | Reachability Analysis for Termination and Confluence of Rewriting TACAS Link to publication |